KR20140039275A - Autonomous locomotion equipment and control method therefor - Google Patents

Abstract

An autonomous moving device comprising a traveling part having wheels driven by a motor, and an upper body part having an environmental sensor for detecting an obstacle in a driving direction, wherein the upper body part includes means for recognizing a position and an obstacle of a magnetic body, Means for evaluating the avoidance capability, means for determining the collision avoidance capability with the obstacle, and means for obtaining the priority of collision avoidance for the passage estimation region of the obstacle from the collision avoidance capability, Is a region where the passage estimation region of the high obstacle does not overlap with the region where the driving unit is located, and moves the collision estimation region of the lower priority obstacle to a range that can avoid collision even if it overlaps with the region where the driving unit exists. A control unit was provided.

Description

Autonomous mobile device and its control method {AUTONOMOUS LOCOMOTION EQUIPMENT AND CONTROL METHOD THEREFOR}

The present invention relates to an autonomous mobile device having an obstacle avoidance capability and a control method thereof.

When an autonomous mobile device such as a robot moves a person's living space, movement of the autonomous mobile device becomes an obstacle for a person in some cases, so that something needs to be dealt with.

As background art in such a case, there exists Unexamined-Japanese-Patent No. 2010-79852 (patent document 1), for example. In this patent document 1, even when an autonomous mobile device moves along a target position track | orbit and the movement is obstructed by an obstacle, the autonomous mobile device can recognize the behavior of an obstacle. In addition, if the autonomous mobile device moves an obstacle which is an autonomous moving object (human being, another autonomous mobile device, etc.), the autonomous mobile device moves along the obstacle if the autonomous mobile device moves along the target position trajectory of the phenomenon. It is possible to control the movement promoting movement.

Japanese Unexamined Patent Application Publication No. 2010-79852 Japanese Unexamined Patent Publication No. 2008-65755

When there are a plurality of obstacles that hinder the movement of the autonomous mobile device, the obstacles may have different ability to accelerate and decelerate or to move in a different direction.

For example, in a hospital, a passage may be blocked at the same time as a normal person, a wheelchair, a disabled person, or the like. In such a case, if the wheelchair or the disabled person has a lower evasion ability than the normal person, and the autonomous mobile device is in a position to block both paths, there is a possibility that the staggering takes time or the two and the autonomous mobile device come into contact with each other. In this case, if the autonomous mobile device moves on the course of the normal person with high avoidance ability, the influence on the movement of the wheelchair or the disabled person is reduced.

However, for example, when the autonomous mobile device moves directly in front of the normal person, since the avoidance behavior of the normal person also has a possibility of time or contact, it is necessary to move to the position where the autonomous mobile device is easily avoided even by the normal person.

As described above, in order to safely and efficiently cope with a situation where the movement of a plurality of obstacles is hindered, the obstacles with low avoidance ability have little influence on the movement of obstacles, and the obstacles having high avoidance ability that can be expected when the autonomous vehicle is avoided. Also, it is necessary to provide an autonomous mobile device which moves to a position where the autonomous mobile device is easy to avoid. However, in the technique described in Patent Literature 1, since the autonomous mobile device does not consider the case where a plurality of obstacles are replaced, there is a possibility that it cannot be coped safely.

An object of the present invention is a position where the autonomous mobile device is easily avoided even when there are a plurality of obstacles that hinder the movement of the autonomous mobile device. It is to provide an autonomous mobile device that can move to.

The above object is an autonomous moving device comprising a traveling part having wheels driven by a motor and an upper body part having an environmental recognition sensor for detecting an obstacle in a driving direction, wherein the upper body part has means for recognizing its position and obstacles. And means for evaluating the avoidance capability of the obstacle, means for determining the collision avoidance capability with the obstacle, and means for obtaining the priority of collision avoidance for the passage estimation region of the obstacle from the collision avoidance capability. The collision estimation region of the high priority obstacle does not overlap the region where the driving portion is located, and the collision can be avoided even if the passage estimation region of the low priority obstacle overlaps with the region where the driving portion exists. It is achieved by having a control unit which moves to a range.

Moreover, it is preferable that the said object calculates the said evaluating ability quantitatively from the speed and the width | variety of the said obstacle.

In addition, it is preferable that the means for evaluating the avoidance capability sets the avoidance priority of the obstacle having the avoidance capability lower than a reference value.

In addition, it is preferable that the means for evaluating the avoidance ability sets the obstacles whose speed is slower than the reference range and the avoidance priority of the obstacle faster than the reference range.

In addition, it is preferable that the above-mentioned objects classify the obstacle into several kinds based on the avoidance ability.

The above object further includes a built-in communication device for communicating with a calculator, which includes an information storage unit, a travel information calculation unit, an obstacle recognition unit, a travel control unit, an obstacle classification unit, a destination point setting unit, A control method of an autonomous mobile device comprising an action determining unit, comprising: determining the presence or absence of information from the obstacle classification unit, and when the information is not sent from the obstacle classification unit, a final destination of the information storage unit is a current destination. Passing the information to the traveling control unit; and if there is information from the obstacle classification unit, obtaining a movable area that is an area where the autonomous mobile device can travel without colliding with the obstacle by the traveling unit; Estimating the passage area of the obstacle, and the area that is virtually movable for calculation. Is achieved by made of an setting a moving area, the method comprising: determining a destination point on the basis of the virtual movement area.

According to the present invention, when there are a plurality of obstacles that hinder the movement of the autonomous device, the influence on the movement of the obstacle with low avoidance ability is small, and even in the obstacle with high avoidance ability, the autonomous device is moved to a position where it is easy to avoid. A mobile autonomous mobile device can be provided.

1 is a conceptual diagram showing the motion of the autonomous mobile device according to the first embodiment of the present invention.
2 is a schematic configuration diagram of an autonomous mobile device according to the first embodiment.
3 is a view for explaining a method of classifying obstacles according to the first embodiment.
4 is an operation flowchart of the destination point determination unit according to the first embodiment.
5 is a view for explaining a movable region according to the first embodiment.
6 is a method for setting an obstacle passing area according to the first embodiment.
7 is a view for explaining a virtual movable area according to the first embodiment.
8 is a view for explaining a method for determining the target point TG according to the first embodiment.
9 is an example of motion when the obstacle according to the first embodiment is only classification A. FIG.
10 is an example of motion when the obstacle according to the first embodiment is only classification B. FIG.
11 is an example of motion when the obstacle according to the first embodiment is only classification C. FIG.
12 is a configuration diagram of an autonomous mobile apparatus according to the second embodiment.
13 is an operation flowchart of the autonomous moving apparatus according to the second embodiment.
FIG. 14 is a diagram for explaining a method for determining a temporary destination TG according to the second embodiment.
FIG. 15 is a view for explaining a method for determining TG in the case where the autonomous mobile apparatus according to the second embodiment cannot pass through the virtual movable area.
FIG. 16 is a view for explaining a method for determining TG according to the third embodiment. FIG.
FIG. 17 is a diagram illustrating an example of setting a passage area matched with the terrain according to the fourth embodiment. FIG.

Hereinafter, a plurality of embodiments will be described with reference to the drawings. In the embodiment, the width of the passageway through which the autonomous vehicle passes is said to be a length that can always be crossed with other obstacles.

Example 1

This embodiment will be described with reference to FIGS. 1 to 11.

An outline of the autonomous mobile device realized by this embodiment will be described with reference to FIG. 1.

1 (a) to 1 (c) are conceptual views showing the movement of the autonomous mobile device according to the first embodiment of the present invention.

In FIG. 1, when the movement is hindered by a plurality of obstacles traveling on the virtual movable area 61 (runway), the autonomous movement apparatus 1 evaluates the avoidance capability of the obstacle and gives an avoidance priority. do.

In the case of FIG. 1A, since the autonomous mobility device 1 has a lower evasion capability of the disabled person B1 and the wheelchair B2 than the normal person C1, the autonomous mobile device 1 recognizes the obstacle as having a high avoidance priority. The areas 51B1 and 51B2 are passage areas inferred that the disabled person B1 and the wheelchair B2 pass forward. The autonomous mobile device 1 which guessed these passing areas 51B1 and 51B2 avoids the disabled person B1 and the wheelchair B2 with high avoidance priority, and is outside the passing areas 51B1 and 51B2, and avoids priority. It moves to the position where the course change of the normal person C1 with low degree is small.

In the state of FIG. 1 (b), the normal person C1 who has confirmed the autonomous mobile device 1 changed to the passage area of the normal person C1 having a low avoidance priority moves to the rear of the disabled person B1 and moves autonomously. Running of the device 1 is prioritized.

In the state of FIG. 1C, after confirming that there is no obstacle in the passage area, the autonomous mobile device 1 can pass through the disabled B1 and the wheelchair B2 as soon as possible.

Under the situation of FIG. 1A, since the conventional autonomous moving apparatus 1 does not evaluate the avoidance capability of an obstacle or set the passage area, it is completely from the passage area | region of the obstacle with low avoidance capability. Do not deviate. For this reason, when the disabled person B1 or the wheelchair B2 with low avoidance ability changes their course, the efficiency is poor or the risk is accompanied.

In contrast, with the autonomous mobile device of the present invention, it is possible to move to an easy-to-avoid position even in an obstacle having high avoidance ability without disturbing the path of the obstacle having a poor avoidance ability.

FIG.2 (a) (b) is a schematic block diagram of the autonomous mobile apparatus which concerns on Example 1. FIG.

Fig. 2 (a) is a mechanical configuration diagram and Fig. 2 (b) is a system configuration diagram of the autonomous mobile apparatus 1 of this embodiment. In the present embodiment, the autonomous mobile device 1 is smaller than a person, and assumes that the vehicle mainly travels indoors.

The mechanical configuration of the present invention will be described with reference to Fig. 2 (a). The autonomous movement apparatus 1 includes a traveling part 11 and an upper body 12. The outside is provided with a calculator 2 connected to the communication device 3, and the autonomous mobile device 1 and the calculator 2 are wirelessly communicated by the communication device 3 and the communication device 124 (indicated by arrows). Information exchange in both directions is performed.

The traveling part 11 is equipped with the wheel 111 and the drive motor 112 provided with the encoder. The upper body 12 includes a battery 121 serving as a power source of the autonomous mobile device and a control device 122. The upper body 12 has a built-in communication device 124 which communicates with the laser scanner and the calculator 2 as an environmental sensor 123 for detecting an obstacle. As shown in FIG. 2 (b), the calculator 2 includes an information storage unit 21, a travel information calculation unit 22, an obstacle recognition unit 23, a travel control unit 24, an obstacle classification unit 25, It consists of the destination point setting unit 26 and the action determination unit 20.

The system configuration part of this invention is demonstrated using FIG. 2 (b). In addition, the detail of the process of each structure part is mentioned later.

First, the obstacle information acquired from the environmental recognition sensor 123 and the wheel rotational angular velocity obtained from the traveling unit 11 are connected to the communication device 124 and the calculator 2 built in the autonomous mobile device 1 ( It is transmitted to the information storage part 21 of the calculator 2 via 3). The information storage unit 21 is calculated by the final destination point input from the behavior determination unit 20, the measurement data from the encoder of the environmental recognition sensor 123 or the motor 112, and the travel information calculation unit 22. Storing information necessary for the control of the autonomous mobile device, such as the current position, direction, and speed of the autonomous mobile device 1, and the current location and speed of the obstacle calculated by the obstacle recognition unit 23, in the memory of the calculator 2, or Update and call data in memory.

The final destination point of the behavior determination unit 20 is determined in advance by a program whether a person inputs from the terminal of the calculator 2. The running information calculation part 22 calculates the present origin reference coordinate and direction, forward speed, and turning speed of the autonomous moving device 1 from the measurement result of the encoder of the motor 112 stored in the information storage part 21. The calculation returns to the information storage unit. The obstacle recognition unit 23 estimates the coordinates, the speed, and the width of the obstacle from the measurement result of the environmental recognition sensor 123 stored in the information storage unit 21, and returns the information to the information storage unit 21.

The travel control unit 24 calculates the target forward speed and the target turning speed based on the obstacle information of the information storage unit 21 and the current destination point set by the destination point setting unit 26. It passes to the control device 122 through. In addition, the travel control unit 24 always determines whether the autonomous mobile device 1 can avoid the obstacle, and if it is impossible to avoid, the running control unit 24 transmits the unavoidable object to the obstacle classification unit 25. If the obstacle classification unit 25 is told that the obstacles cannot be avoided from the running control unit 24, the obstacle classification unit 25 classifies the obstacles by the avoidance capabilities based on the obstacle information of the information storage unit 21, and classifies the avoidance priority of the obstacles. The determination is made and sent to the destination point setting unit 26.

When the information is not sent from the obstacle classification unit 25, the destination point setting unit 26 transmits the last destination point of the travel information storage 21 to the travel control unit 24 as the current destination point, and the obstacle. When the information is sent from the classification unit 25, the obstacles with high avoidance priority are not disturbed, and the obstacles with low avoidance priority easily calculate the point where the autonomous moving object 1 is avoided, It conveys to the traveling control part 24 as a destination point.

After that, the traveling control unit 24 calculates the straight forward speed and the turning speed for advancing while avoiding the obstacle to the destination point set by the destination point setting unit 26, and controls the control device (through the communication devices 3 and 124). 122). The control apparatus 122 controls the movement direction and the movement speed of the travel part 11 based on the speed instruction from the information storage part 21 and the information of the travel part 11.

Hereinafter, the detail of the calculation process of each structure part is demonstrated.

In this embodiment, a laser scanner is used as the environmental recognition sensor 123, and the data string of the distance to the obstacle is transmitted to the information storage unit 21 via the communication devices 124 and 3 at predetermined angular intervals. In addition, the encoder of the motor 112 detects the rotational angular velocity of the wheel and transmits it to the information storage unit 21 via the communication devices 124 and 3.

The information storage unit 21 stores the origin set by the user at the time of autonomous moving object, the coordinates of the final destination point based on the origin, obstacle information for the past few seconds, the rotational angular velocity of the wheel, the travel information calculation unit 22 and the obstacle recognition. The calculation result of the section 23 is stored in the memory of the calculation section 2.

The running information calculation part 22 is based on the rotational angular velocity of the wheel memorize | stored in the information storage part 21, and the visual force of the direction of the autonomous moving device 1, and the present origin of the autonomous moving device 1 is carried out. Reference coordinates and directions, forward speeds and swing speeds are calculated and returned to the information storage section 21 again.

The obstacle recognition unit 23 estimates the coordinates, the speed, and the width of the obstacle from the data obtained from the environmental recognition sensor 123, and transmits the obstacles to the information storage unit 21. Although the laser scanner is used in the present embodiment, since the data obtained is a data string at predetermined angular intervals, a recognition method for recognizing a plurality of obstacles as respective obstacles is required.

For example, there is a method of Japanese Patent Laid-Open No. 2008-65755 as a recognition method.

In this method, first, a sudden change point with respect to an angle of a distance value obtained from a laser scanner at a certain time t is detected, the data string is divided into groups of consecutive points, and the information storage unit 21 is stored as a segment. Preserve Thereby, the characteristic quantity of representative positions, such as the center of each segment, shape, etc. in time t is recognized. Next, similar calculation is performed at time t + Δt to obtain a feature amount of each segment.

Here, the characteristic amount of the segment obtained at the time t and the characteristic amount of the segment obtained at the time t + Δt are compared, and the segments having the closest feature amounts are recognized as the same obstacles, and the speed of the obstacle is changed from the change amount of the representative position. The width can be obtained from the shape. In addition, the obstacle having a movement speed of approximately 0 is regarded as a stationary obstacle, and each data point by the laser scanner is regarded as an obstacle of width 0 and stored in the information storage unit 21.

In the traveling control unit 24, a generally known obstacle avoidance method "The Dynamic Window Approach to Collision Avoidance." (IEEE Robotics & Automation Magazine 4 (1), pp. 23-33, 1997) is used. In this method, the target turning speed candidates (p1, p2, ..., pk) of several stages that the autonomous mobile device 1 can travel, and the target advance speed candidates (v1, v2, ... vq) of several steps are possible. Evaluate distance function with obstacle (Lcol (pk, vq)), direction function (θgoal (pk)) and forward velocity function (V (vq)) for the destination point coordinates sent from the destination point setting unit in the case of going forward. Where the objective function shown in Equation 1 is obtained by adding the weights (α, β, and γ) to

Choose pk, vq that makes G (pk, vq) larger. α, β, and γ can be set by simulation or empirical rules. The selected target turning speed pk and the target forward speed vq are transmitted to the control device 122 via the communication devices 3 and 124. If the angle (θ) of the velocity vector with respect to the target point can only be selected as a moving destination candidate of 90 ° or more, or when G (pk, vq) becomes maximum when vq ≤ 0, avoiding approaching the target point It is judged that it is impossible, and a flag for avoiding is sent to the obstacle classification unit 23.

When the obstacle classification unit 25 receives the non-avoidance flag from the traveling control unit, the obstacle classification unit 25 classifies each obstacle by the avoidance capability based on the speed and the width of the obstacle stored in the information storage unit 21, and classifies the avoidance priority. Decide A healthy pedestrian is said to have the highest evasion ability. If the width is larger than that of a general pedestrian, it is said to be a mobile body other than human, and the evasion ability is low. In addition, a moving object having a slower moving speed than a healthy pedestrian is a person other than a person, a person carrying a heavy thing, a disabled person, etc. A moving object having a faster speed than a healthy pedestrian is a sudden speed change such as a running person or a bicycle. In other words, it is said that it is difficult to change course and that evasion ability is low. In this embodiment, the sum (R) of both weights (a, b) over the width (W) and velocity (V) of the obstacle.

Evaluate the obstacle avoidance ability, classify the obstacle from R, the width (W), and the speed (V) of the obstacle. The weights a and b are calculated | required from the upper limit value Rmax of R of the normal person mentioned later. The classification method will be described with reference to FIG. 3.

3 is a view for explaining a method of classifying obstacles according to the first embodiment.

In FIG. 3, when the lower limit speed determined as a healthy pedestrian is Vmin and the upper limit width is Wmax and the upper limit of R is Rmax,

It is classified into A, and the avoidance priority is set to A> B> C. Rmax is determined as a heuristic so that it can be classified as shown in FIG. Moreover, a and b of Formula 2 are determined from Rmax. Like the classification A, since the large moving object is a large and heavy moving object or a moving object moving at high speed, it is difficult to change the speed or change the course, and the autonomous mobile device 1 needs to actively avoid it. As in the classification B, a moving object having small V and R and having a large W is difficult to change the course from the slowness of its size and speed, but it is easy to stop, so that the collision with the autonomous mobile device 1 is stopped. It can be inferred that this can be avoided. The classification C has a level of evasion capability that is normal from the speed and the width, and it can be estimated that the evasion of the autonomous mobile device 1 is easy.

The operation of the destination point setting unit 26 will be described with reference to FIGS. 5 to 8 based on the flowchart of FIG. 4.

4 is an operation flowchart of the destination point determination unit according to the first embodiment.

In FIG. 4, when no information is sent from the obstacle classification unit 25 in S100, the flow advances to S101. When the information is sent from the obstacle classification unit 25 to the destination point setting unit 26, the process proceeds to S102 and does not disturb the path of the obstacle having a high avoidance priority, and furthermore, the obstacle having a low avoidance priority is an autonomous moving object. (1) The point TG which is easy to avoid is calculated, and it conveys to the traveling control part 24, using TG as an objective point.

In S101, the target point setting unit 26 transmits the final destination point of the information storage unit 21 to the travel control unit 24 as the current destination point TG.

In S102, as shown in an oblique line in FIG. 5, the movable portion 60 obtains the movable region 60, which is an region in which the autonomous mobile apparatus 1 can travel without colliding with an obstacle. Equation 4 is obtained from the coordinates Oi and the width Wi of the i-th obstacle (i = 1, 2, ..., N) stored in the information storage unit 21 with Pa as the movable region 60. define.

Here, r is the distance necessary for the obstacle to safely avoid collision with the autonomous mobile device 1, the size of the autonomous mobile device 1, the interval when crossing with other moving objects, and another configuration of the autonomous mobile device 1 When operating a part, set it based on the surroundings and spacing.

In S103, the passage area 51 of the obstacle currently captured is estimated. When the coordinates of the origin centers of the N obstacles Oi obtained from the information storage unit 21 are Oi, the velocity vector is Vi, and the unit normal vector of Vi is ni, the passage region T (Oi) of Oi is expressed by Equation 5 and Estimate together.

As shown in Fig. 6, when the current position Xi of the autonomous mobile device 1 is substituted on the left side of Equation 5, and s and t are found, the values of s and t are s≥0 and | t | ≤Wi When / 2 + r is satisfied, it is determined that the autonomous mobile device 1 enters the passage region of the obstacle Oi.

In S104, the virtual movable area 61 (Fig. 7 oblique part) which is an area set as virtually movable for calculation is set. The virtual movable area 61 is set as follows based on the avoidance priority set by the obstacle classification unit 25.

(1) When N kinds of obstacles Ai (i = 1, 2, ..., N) of the movable area 60 or the virtual movable area 61 are only A, as shown in FIG. The passage area 51Ai is made impossible to pass, and the remaining area is made into the virtual movable area 61P. When the coordinates of Ai's origin reference point are Oi, the width is Wi, the velocity vector of each obstacle is Vi, and the unit normal vector of Vi is ni, the virtual movable region 61P becomes equation (6).

(2) When N kinds of obstacles Bi (i = 1, 2, ..., N) of the movable region 60 or the virtual movable region 61 are only B, as shown in FIG. 7 (b), The passage area 51Bi of the obstacle Bi (i ≠ k) other than the slowest obstacle Bk is made impossible to move, and the remaining area is made the virtual movable area 61P (Fig. 7 (b)) ).

Moreover, as said Bk, you may select the obstacle with the longest time to collision.

(3) When the obstacle of the movable area 60 or the virtual movable area 61 is C only, the current movable area 60 or the virtual movable area is used as it is.

(4) When at least two or more kinds of obstacles among A, B, and C exist in the movable area 60, the passage area of all obstacles other than the obstacles having the lowest avoidance priority is made impossible to move, and the remaining area. Denotes the virtual movable area 61. As shown in Fig. 7 (c-1), the P obstacles Mi (i = 1, 2, ..., P) are the obstacles having the lowest priority, and the Q obstacles Nj (j = 1, 2, ...). If Q) is another kind of obstacle, the coordinates of Nj are Nj, the velocity is Vj, and the unit normal vector of Vj is nj, the virtual movable region 61P (the oblique portion in Fig. 7 (c-1)) is (8)

By setting the virtual movable area 61 as described above, the obstacles present in the virtual movable area 61 can be limited to one type of B or C, so that the above 2) or 3) is performed again and the virtual By updating the movable area, the virtual movable area 61 finally becomes as shown in Fig. 7C-2.

In S105, the destination point TG is determined based on the virtual movable area 61 set in S104. The determination method of TG is demonstrated using FIG.

When the types of obstacles are all class A, the advancing direction of the autonomous movement apparatus 1 is blocked in the area | region which cannot be moved like FIG.8 (a). In this case, the point in the virtual movable area closest to the rear of the autonomous movement apparatus 1 is made into the target point TG. If the obstacle includes a obstacle other than the classification A, only one obstacle of the classification B or C exists in the virtual movable area 61 by the processing of 2) 3) 4) of S104, or A plurality of obstacles exist.

If only one obstacle O5 exists in the virtual movable region 61 as shown in FIG. 8B, when the autonomous mobile apparatus 1 moves to the boundary line L1 or L2 of the virtual movable region 61, The course change of the obstacle O5 is small. In order to determine the position on L1 and L2 entirely, for example, the intersection of the stop position 71 and L1, L2 when the autonomous movement device 1 decelerates at the maximum acceleration in the possible turning direction as a candidate for TG. If the distance between the autonomous mobile device 1 and L1 is greater than the center of the autonomous mobile device 1 and the distance between L2, the intersection of L1 and the stop position 71 is set as the temporary destination TG. The intersection of the stop position 71 is made into the temporary destination TG. In addition, when there are N obstacles (Oi, i = 1, 2, ..., N) in the virtual movable area 61 as shown in FIG. 8C, the boundary line L1 of the virtual movable area 61 is present. , L2) and the middle line (Lik) of the boundary lines of the passing areas (51Oi, 51Ok) of neighboring obstacles (Oi and Ok) (k = 1, 2, N, | ik | = 1) When the autonomous platform 1 moves, the path change of the obstacles is reduced. Therefore, among the intersections of the stop position 71 and the L3, L4, and Lik, the passage area of the obstacles is the same as in the case of one obstacle. The point where the boundary and the distance between the autonomous mobile device 1 are large is set as the temporary destination point TG.

The control apparatus 122 controls the motor 112 of the traveling part 11 so that the forward speed and the turning speed of the autonomous mobile device 1 may be performed as instruct | indicated by the travel control part 24. FIG.

Hereinafter, the operation example of the autonomous mobile apparatus 1 is demonstrated using FIG. 1, FIG. 9, FIG. 10, and FIG.

In these figures, in order to make a concept easy to understand, r in Formula 4-Formula 8 is set to 0, Instead, the occupied area | region 72 of the autonomous vehicle 1 is replaced by the radius r and the center of the autonomous vehicle. It is shown by the circle which made the position.

9 is an example of motion when the obstacle according to the first embodiment was only classification A. FIG.

10 is an example of motion when the obstacle according to the first embodiment was only classification B. FIG.

11 is an example of motion when the obstacle according to the first embodiment was only classification C. FIG.

1 is an example in the case where the movement of the autonomous movement device 1 is disturbed by a plurality of kinds of obstacles.

For example, a normal person C1 is classified as C, and a disabled person B1 and a wheelchair B2 are classified as B. Since normal person C1 of class C is the kind with the lowest avoidance priority, as shown to FIG. 1 (a), the autonomous mobile apparatus 1 passes through obstacles B1 and B2 other than normal person C1. Becomes impossible to pass, and the virtual passage area 61 is set. When the destination point TG is determined in accordance with S105 of Fig. 4, the TG does not disturb the course of B1 and B2, and the course change of C1 is set at a position where there is little.

When the autonomous mobile device moves to TG as shown in Fig. 1 (b), it can be expected that B1 and B2 proceed as they are, and C1 takes an evasive action. As shown in Fig. 1 (3), the autonomous movement apparatus 1 becomes movable according to the avoidance behavior of C1, and starts moving to the final destination point again.

9 is an example in the case where the movement of the autonomous mobile device 1 is disturbed by the bed A1 and the side walkers A2 and A3 during the emergency conveyance.

In FIG. 9, A1, A2, and A3 move faster than the equation 3 Vmax together and move at the same speed, resulting in class A. In FIG. Therefore, the passage area | region of all obstacles becomes impossible to move, and the virtual movable area 61 becomes as shown to Fig.9 (a). Since there is no virtual movable area 61 containing the occupied area 72 at a position where the autonomous mobile device 1 can stop at the maximum acceleration, the temporary destination point TG is the autonomous mobile device 1. It is set at the rear of, and as shown in FIG. 9 (b), the autonomous movement device 1 moves to a temporary destination point TG. In this way, the autonomous moving device 1 can wait for the passage of the obstacle by TG without affecting the movement of the obstacle. After all obstacles pass as shown in Fig. 9C, the autonomous movement device 1 starts moving toward the final destination point again.

10 is an example in which the movement of the autonomous mobile device 1 is blocked by the disabled B1 and the wheelchair B2.

In FIG. 10, the slow-disabled person B1 and the large wheelchair B2 are classified B together. When the movement speed is slower for the disabled B1, the passage area 51B2 of the obstacle B2 other than the B1 becomes impassable, as shown in Fig. 10A. Within the set virtual moving area 61, the temporary destination point TG is set so that the occupied area 72 does not overlap 51B2, and as shown in FIG. 10 (b), the autonomous mobile device 1 is a temporary destination point TG. Go to). Since the disabled B1 is moving along the wall, it is expected that the course is difficult to change, but since the speed of B1 is slow, it is easy to stop, and the wheelchair B2 proceeds as shown in FIG. It can be expected that the autonomous mobile device 1 can pass by allowing the disabled B1 to wait until there is a distance that can pass between the autonomous mobile device 1 and the autonomous mobile device 1. When the autonomous mobile device 1 becomes movable, the autonomous mobile device 1 starts moving toward the final destination point again.

11 is a case where the autonomous mobile device 1 is blocked by three pedestrians in the passage.

In FIG. 11, since the pedestrians C1, C2, and C3 become the power source classification C, the virtual movable region 61 becomes the same as the movable region 60 as shown in FIG. 11A. By S105 shown in FIG. 4, TG is set on the midline of the boundary lines of the passage areas 51C1 and 51C2 of each pedestrian C1 and C2, and when the autonomous mobile device 1 moves to TG, FIG. It can be expected that the pedestrians C1 and C2 take an evasive action as shown in FIG. When the space between the pedestrians C1 and C2 is opened and the autonomous mobile device 1 is movable, the autonomous mobile device 1 starts moving toward the final destination point as shown in Fig. 11C again.

As described above, when the obstacle is blocked by the obstacle, the autonomous moving apparatus 1 classifies the obstacle by the avoidance ability, gives the avoidance priority, and estimates the passage area of the obstacle, thereby invading the passage area of the moving object having the low avoidance capability. Furthermore, since the vehicle moves to a position where the path change of the moving object having high avoidance ability is less, the collision between the obstacle and the autonomous moving object can be avoided safely and efficiently.

Example 2

In the present embodiment, the case where the autonomous mobile device itself is provided with a calculator will be described with reference to Figs.

The autonomous mobile device of the present embodiment is large enough to be carried by a person, and mainly travels outdoors. The present Example and Example 1 are different from the point described below, and since the other point is basically the same as Example 1, the overlapping description is abbreviate | omitted. In the present embodiment, the same effect is obtained in the configuration common to that of the first embodiment.

12 is a configuration diagram of an autonomous mobile apparatus according to the second embodiment.

13 is an operation flowchart of the autonomous mobile apparatus according to the second embodiment.

FIG. 14 is a diagram for explaining a method for determining a temporary destination TG according to the second embodiment.

FIG. 15 is a view for explaining a method of determining TG in the case where the autonomous mobile apparatus according to the second embodiment cannot pass through the virtual movable area.

In FIG. 12, the mechanism block diagram of the autonomous movement apparatus 8 of a present Example is shown to FIG. 12 (a), and a system block diagram is shown to FIG. 12 (b). The autonomous mobile device 8 consists of a travel part 81 and a boarding part 82. In the first embodiment, the calculator was external, but in the present embodiment, the calculator 9 is provided in the traveling section 81, and the control device 814 provided in the traveling section 81 is wired with the calculator 9 in a wired manner. You are connected. Therefore, in the first embodiment, components that have been transmitting and receiving information through the communicators 3 and 124 can send and receive information directly by wire. In addition, the driving unit 81 includes a wheel 811 for moving the autonomous moving device, a driving motor 812 having an encoder independently of the rear left and right wheels, and a battery 813 serving as a power source. do. The board | substrate 82 is a quantity of content which a person enters, and is equipped with the laser scanner as the environmental recognition sensor 821. As shown in FIG.

As shown in FIG. 13, in Example 2, the area | region occupied by the autonomous movement apparatus 8 is represented by the rectangle which made width of the progress report into r of Formula 4, The distance between the obstacle and the surface of the autonomous vehicle along the direction can also be estimated. In addition, the length of the front-back direction of the occupied area 71 is set based on the magnitude | size of the autonomous movement apparatus 8, the space | interval, and the space | interval created when the autonomous movement apparatus 8 stops.

The traveling control unit 94 controls such that obstacles which are different in direction as well as the position of the autonomous mobile device 8 are easily avoided. For example, the angle function θgoal in Equation 1 is parallel to the boundary line of the passage region of the obstacle, as shown in Fig. 14 (a) (when the obstacle in the virtual movable region is singular), or Fig. 14 (b). As in the case where there are a plurality of obstacles in the virtually movable region, the path change of the staggered obstacles can be reduced by setting the values so as to be larger in the direction parallel to the center line of the boundary lines of adjacent passage regions.

Since the autonomous mobile device 8 of the present embodiment assumes that a person is burned, the width r of the occupied area 71 is larger than that of the person. Therefore, when the virtual movable area 61 is set similarly to the first embodiment, as shown in FIG. 15, the occupied area 72 is completely separated from the passage area 51 of the obstacle which becomes impossible to move. You may not be able to.

Only in FIG. 15 (when the virtual movable region is narrower than the width of the autonomous mobile device), for the sake of easy understanding, the virtual movable region 61 is shown ignoring the width r of the occupied region 72. have. If the occupied region 72 cannot be completely removed from the passage region 51 of the obstruction that has become non-movable, R in the order in which R is not movable among the passage regions 51 of the obstacle other than the classification A. It is possible to move one by one, and try to reset so that the occupied area 72 does not enter the passage area 51 which cannot pass. If it is impossible to reset or the obstacle of the pass-through area which cannot be moved is only classification A, the autonomous mover (in the virtual pass-through area 61 behind the autonomous mover 8 as in the first embodiment) In 61), the shortest point is set as the target point TG.

Example 3

In Examples 1 and 2, the temporary destination point TG has obstacles C1 and L2 when the center line of the current passage area of the obstacles C1 and C2 is avoided as shown in FIG. 16. When the center line of the passage region of C2) is set to L1 'and L2', the TG is a position where the angle (θ1) formed by L1 and L1 'and the angle (θ2) formed by L2 and L2' are the same. It is possible to reduce the course change of both C1 and C2).

Example 4

In Examples 1 and 2, the obstacle passing-through region 51 may be set not only in accordance with the speed and width of the obstacle, but also in accordance with the movement pattern of the obstacle and the surrounding terrain. For example, as shown in Fig. 17, if the curve is an obstacle during passage, a method of setting the circular arc-shaped passage region in accordance with the curve, rather than the linear passage region, is considered.

Example 5

In Examples 1 to 4, the traveling control unit can be suitably used as long as it is a method capable of controlling the autonomous mobile device in an environment in which an obstacle exists. For example, if it is a method of conveying a target speed to a control apparatus, it will determine that obstacle avoidance is impossible, for example, when the target speed becomes 0 or less. For example, when the traveling control section generates the route to the destination, when the traveling control section judges that the route cannot be generated, it can be determined to be impossible to avoid. In addition, if the traveling part is an apparatus for moving an autonomous moving device, a wheel, a leg, a hover, etc. provided with a steering mechanism can be used suitably according to a surrounding environment.

The environmental recognition unit is capable of estimating the width, position, and speed of an obstacle, and is a communication device that detects various sensors using millimeter waves, ultrasonic waves, pressure sensors, and the like when an obstacle to be avoided has an IC tag. Or a combination thereof can be suitably used.

In the obstacle classification unit, a transmitter is provided for each autonomous mobile device or obstacle, and the type, position, and speed information of the obstacle are received by a receiver installed near the autonomous mobile device or the autonomous mobile device and transmitted to the calculator. The obstacle may be classified without evaluating the obstacle avoidance ability. Alternatively, the obstacle may be classified from the appearance of the obstacle by using a camera, a 3D laser scanner, or the like for acquiring the appearance. In addition, you may use weight instead of the width | variety of an obstacle. In the case of using the weight, a pressure sensor is placed on the entire floor surface of the autonomous moving device, and the weight of the obstacle can be estimated from the measurement result. In addition, the avoidance priority may be set in consideration of the moving place of the autonomous mobile device. For example, the autonomous mobile device moves based on the map data, and the map data maintains the avoidance priority setting method for each area, and there is a method of setting the avoidance priority based thereon.

1: Autonomous moving apparatus of Example 1 11: Running part of Example 1
111: Wheel of Example 1 112: Motor of Example 1
12: Upper Body of Example 1 121: Battery of Example 1
122: control apparatus of Example 1
123: Environmental Awareness Sensor of Example 1
124: a communicator embedded in the upper body portion 11
2: calculator of Example 1
20: Behavior determination unit of Example 1
21: Information storage section of Example 1
22: driving information calculation unit of the first embodiment
23: obstacle recognition unit of Example 1
24: Travel control unit of Example 1
25: Obstruction classification part of Example 1
26: destination point setter of Example 1
3: communicator 51 connected to calculator 2: passing area
60: movable area 61: virtual movable area
71: shortest stop position of the autonomous vehicle
72: occupied area of autonomous vehicle
8: Autonomous Mobile Device of Example 2 81: Traveling Part of Example 2
811: Wheel of Example 2 812: Motor of Example 2
813: the battery of Example 2
814: Control apparatus of example 2 82: boarding portion of example 2
821: Environmental Awareness Sensor of Example 2 9: Calculator of Example 2
91: information storage section 2
92: driving information calculation unit of the second embodiment
93: obstacle recognition unit of Example 2
94: driving control unit of Example 2
95: obstacle classification unit of Example 2
96: destination point setter of Example 2
Oi, Ai, Bi, Ci: i-th obstacle
Li: boundary line of the i-th virtual movable area 61
Ljk: the middle line of the boundary line between the jth passing area 51 and the kth passing area 51 which are adjacent to each other
TG: temporary destination

Claims (6)

In the autonomous mobile device which consists of a travel part which has a wheel driven by a motor, and an upper body part which has an environmental recognition sensor which detects the obstacle of a travel direction,
The upper body portion includes means for recognizing its position and obstacles, means for evaluating the avoidance ability of the obstacles, means for determining the collision avoidance ability with the obstacles, and an area for estimating the passage of the obstacles from the collision avoidance ability. A means for obtaining a priority of collision avoidance is provided.
Range where the passage estimation region of the high priority obstacle does not overlap the region where the traveling portion is located, and the collision can be avoided even if the passage estimation region of the low priority obstacle overlaps with the region where the driving portion exists. And a control unit for moving the control unit. The method according to claim 1,
And means for evaluating said avoidance capability is calculated quantitatively from the speed and width of said obstacle. The method according to claim 1,
And the means for evaluating the avoidance capability sets the avoidance priority of the obstacle with the avoidance capability lower than a reference value. 4. The method according to any one of claims 1 to 3,
And the means for evaluating the avoidance ability sets the avoidance priority of the obstacle slower than the reference range and the obstacle faster than the reference range. 4. The method according to any one of claims 1 to 3,
The means for evaluating the evasion capability classifies the obstacle into several kinds based on the evasion capability. Built-in communication device for communicating with the calculator, the calculator comprises an information storage unit, a driving information calculation unit, an obstacle recognition unit, a driving control unit, an obstacle classification unit, an object point setting unit, and an action determining unit In the control method of a mobile device,
Determining the presence or absence of information from the obstacle classification unit;
When information is not sent from the obstacle classification unit, passing the final destination of the information storage unit as the current destination to the traveling control unit;
If there is information from the obstacle sorting unit, obtaining a movable area which is an area in which the autonomous mobile device can travel without colliding with the obstacle by the driving unit;
Estimating the passage area of the currently captured obstacle;
Setting up a virtual movable area that is an area that is virtually movable for calculation;
And determining a destination point based on the virtual movable area.

Images